Conditional random field-recurrent neural network segmentation with optimized deep learning for brain tumour classification using magnetic resonance imaging

ABSTRACT

Brain tumour is one of the most dreadful kinds of tumour caused by the uncontrolled increase in the cells inside the skull. Clinically, brain tumour is diagnosed using different imaging methods, among which Magnetic Resonance Imaging (MRI) is the most extensively utilized technique. The proposed work intends to develop a Deep Learning (DL) based approach for segmenting and classifying to diagnose brain tumours from the brain MRI. This work proposes two novel approaches: (i) tumour segmentation technique using Conditional Random Field-Recurrent Neural Network (CRF-RNN), whose weights are adapted with the help of the novel Chronological Artificial Hummingbird Algorithm (CAHA). (ii) A brain tumour classification scheme using LeNet. Further, an optimization algorithm called Chronological Artificial Vultures Optimization (CAVO) is proposed for the weight optimization of the LeNet. It is observed that the devised model attained higher values of specificity, sensitivity, and accuracy of 0.938, 0.941, and 0.930.

KEYWORDS:

• Brain tumour
• segmentation
• classification
• deep learning
• optimization
• malignant
• Recurrent Neural Network
• Magnetic Resonance Imaging

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Geetha M

Dr. Geetha M Working as Professor in the Department of Computer Science and Engineering, Chennai Institute of Technology, Chennai, Tamil Nadu, India, having 24 Years of experience in Teaching, and has obtained her under graduate degree B.E from Bharathidasan University, Tamil Nadu, India, and Post graduate degree M.E from Satyabhama Institute of Science and Technology. Chennai, Tamil Nadu, India. She completed her Ph.D. degree from Anna University, Chennai, Tamil Nadu, India, in the area of Pervasive Computing. Her main research interests are Automata Theory, Information Security, Machine Learning and Computational Linguistics. Apart from these, her research interests also include Edge Computing and Internet of Things. She has published her papers in more than 50 refereed Journals and also written a book chapter “Biomedical Data Mining for Improved Clinical Diagnosis” in Artificial Intelligence in Data Mining, Elsevier Inc, 2021.

Prasanna Lakshmi K

Dr Prasanna Lakshmi K, earned her Doctorate in Computer Science and Engineering from Jawaharlal Nehru Technological University, Hyderabad, Master Degree in Computer Science and Engineering from Osmania University, Hyderabad. She is currently working as Professor in Information Technology Department, Gokaraju Rangaraju Institute of Engineering and Technology, Hyderabad. She has more than 20 years of Teaching Experience and 4 years of Research Experience. Published 32 research publications in reputed journals which are indexed in Web of Science and Scopus. Guided 9 PG projects and 15 UG projects. She published 1 Indian Patent and been granted with 1 Australian Patent. An active member in Advisory Board Committee for many International Conferences organized through IEEE. Reviewed more than 100 manuscripts being a member of Technical Review Committee in various renowned journals and conferences.Awarded with Global Eminent Teacher award in 2021 and with 50 Fabulous EduTech Leaders award in 2019. She has professional memberships from IACSIT, IAENG and CSTA. Apart from research activities, she also served the institution in many administrative roles like Member Secretary for Academic Council, Chairman Board of Studies, Head of Information Technology Department and Dean of Faculty Development in GRIET. Currently serving Institution with Dean Academics role. Implemented OBE based teaching in classroom. Her research interests include Data Stream Mining, Machine Learning, Natural Language Processing, IoT and Social Networking.

Sajeev Ram Arumugam

Prof. Sajeev Ram Arumugam is currently associated with Sri Krishna College of Engineering and Technology as Associate Professor in Department of Artificial Intelligence and Data Science. He was graduated in B.E.ECE in 2007 followed by M,E CSE in the year 2011 and completed his Ph.D from Vels Institute of Science, Technology & Advanced Studiesin 2019. Prof. Sajeev have authored more than 20+ research articles and Book Chapters in the field of medical image processing, Deep Learning, IoT. He have also participated in 10+ national and international conferences. He is also associated with many journals and have peer reviewed 10+articles for 3 journals. He have authored 2 books, and his research interest includes developing Computer Aided Detection systems in medical fields using Machine and Deep Learning Algorithms, and Developing IoT based AI systems in various fields.

Sandhya N

Dr. Sandhya N currently works at VNR Vignana Jyothi Institute of Engineering and Technology. Her research area is Data Science. She has published around 30 papers in reputed international journals and presented papers at international conferences. She has an experience of 18 years in teaching and administration. Her scholars are working on text mining, big data analytics and sentiment mining.